Re: [quicwg/base-drafts] Encrypting Retry token (#3274)

Nick Banks <notifications@github.com> Fri, 06 December 2019 17:07 UTC

Return-Path: <bounces+848413-a050-quic-issues=ietf.org@sgmail.github.com>
X-Original-To: quic-issues@ietfa.amsl.com
Delivered-To: quic-issues@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4E1471200CD for <quic-issues@ietfa.amsl.com>; Fri, 6 Dec 2019 09:07:37 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3
X-Spam-Level:
X-Spam-Status: No, score=-3 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, MAILING_LIST_MULTI=-1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=github.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iXtXMKA3CuGF for <quic-issues@ietfa.amsl.com>; Fri, 6 Dec 2019 09:07:35 -0800 (PST)
Received: from o7.sgmail.github.com (o7.sgmail.github.com [167.89.101.198]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BB9AB12007C for <quic-issues@ietf.org>; Fri, 6 Dec 2019 09:07:35 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=github.com; h=from:reply-to:to:cc:in-reply-to:references:subject:mime-version:content-type:content-transfer-encoding:list-id:list-archive:list-post:list-unsubscribe; s=s20150108; bh=yFHAqXV2T0YroW7H/9earCyMJlJlQqgTYicfFrk9Sfs=; b= bxdvzQXPOnDopZkyDM5GfSPK9AlzW3JQr+BEQH6RUN1vJB++8vbeEUJjBG4QolaZ frKBaTjeDdNq/ejgx84Wl2hCY2U2NjTczXqfRmaFHtmM7DAkJ3VbTOnCvoEtrOWe xX4ztMTxNIuov3NITS0KJFZZK87z7WHRko61okV4uR0=
Received: by filter0953p1las1.sendgrid.net with SMTP id filter0953p1las1-14355-5DEA8AD5-55 2019-12-06 17:07:33.992997324 +0000 UTC m=+160750.157668040
Received: from github-lowworker-b40b5a4.cp1-iad.github.net (unknown [140.82.115.10]) by ismtpd0016p1iad2.sendgrid.net (SG) with ESMTP id oEoe0vWAQ66wYmr4QD-kgQ for <quic-issues@ietf.org>; Fri, 06 Dec 2019 17:07:33.919 +0000 (UTC)
Received: from github.com (localhost [127.0.0.1]) by github-lowworker-b40b5a4.cp1-iad.github.net (Postfix) with ESMTP id D3FB63E09B9 for <quic-issues@ietf.org>; Fri, 6 Dec 2019 09:07:33 -0800 (PST)
Date: Fri, 06 Dec 2019 17:07:34 +0000 (UTC)
From: Nick Banks <notifications@github.com>
Reply-To: quicwg/base-drafts <reply+AFTOJK3BIPYQ7EANCSGRBY53666VLEVBNHHB7CUNWA@reply.github.com>
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Message-ID: <quicwg/base-drafts/issues/3274/562656995@github.com>
In-Reply-To: <quicwg/base-drafts/issues/3274@github.com>
References: <quicwg/base-drafts/issues/3274@github.com>
Subject: Re: [quicwg/base-drafts] Encrypting Retry token (#3274)
Mime-Version: 1.0
Content-Type: multipart/alternative; boundary="--==_mimepart_5dea8ad5d2a11_72513fd747ecd96c979a8"; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: list
X-GitHub-Sender: nibanks
X-GitHub-Recipient: quic-issues
X-GitHub-Reason: subscribed
X-Auto-Response-Suppress: All
X-GitHub-Recipient-Address: quic-issues@ietf.org
X-SG-EID: l64QuQ2uJCcEyUykJbxN122A6QRmEpucztpreh3Pak3uhj9DlBklxSYysIp6Nnf2MEpy2kW/0ilYjz f1rA6Yu3WY5vEzC1zslO/96CKJ0XHG69Us+wiYlzhTKZRVtcsIVE3+KO/iY+/rPk4Uez56Lzq99070 ieHTX75JPPFnTyK0rw/gMHBCCLjoeIktAxl8DFcKy/tJgDiptMgEU7fAHX0M3ZffpTKkJTxjkTzOnj c=
Archived-At: <https://mailarchive.ietf.org/arch/msg/quic-issues/5luRMOMwYFGWhOtjuLnbTCEMm0s>
X-BeenThere: quic-issues@ietf.org
X-Mailman-Version: 2.1.29
List-Id: Notification list for GitHub issues related to the QUIC WG <quic-issues.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/quic-issues>, <mailto:quic-issues-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/quic-issues/>
List-Post: <mailto:quic-issues@ietf.org>
List-Help: <mailto:quic-issues-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/quic-issues>, <mailto:quic-issues-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Dec 2019 17:07:37 -0000

Here's a quote from our hardware partner:

> Here's some preliminary numbers. Due to code and library complexities, my team could not do the AES testing, but we could generate some results with the ghash.
> 
> The main thing is, encryption vs no encryption.
> Using no encryption as base (1x),  with ghash calculated, the CPU usage is roughly ~2.5x higher.
> 
> And if we expect AES to be 1.7x to 2.8x slower than ghash, we're looking at 4.25x to 7x slower for no encryption vs AES encryption.

And a bit more details in follow up:

> Again, these are preliminary numbers, and since we use ghash as benchmark base, the factor with different cpu generation & AES support is a little bit reduced here...
> 
> And, of course, the code is not tuned to the optimal, as we just took from the libraries mentioned by the link. (And I don't know how much of room left to tune the ghash & aes)
> 
> Attached is a little bit more details:
```
No ghash
------------------------------------------------------------
CPU Usage
Data1   14%   12%   11%   14%   12%

Port   InPPS   InBPS   OutPPS   OutBPS   
--------------------------------------------------------------------------
1   21738.64   22651666.98   0.00   0.00  
2   0.40   23.98   21738.64   22651666.98  

Ghash
------------------------------------------------------------
Data1   21%   29%   26%   27%   30%

Port   InPPS   InBPS   OutPPS   OutBPS  
--------------------------------------------------------------------------
1   20850.42   21725742.72   0.40   23.98   
2   0.40   23.98   20850.02   21725718.74  
```

I'm still of the opinion that, since the Retry packet is almost purely a DoS mitigation mechanism, our view should be what's the absolute simplest design we can take and still get what's absolutely necessary, not nice to have; instead of what else can we squeeze into this feature and hopefully not make it "too expensive".

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/3274#issuecomment-562656995